Rewinders are used to convert large parent rolls of paper into retail sized rolls and bathroom tissue and paper towels. Two types of rewinders are commonly used—center rewinders and surface rewinders. Center rewinders are described, for example, in U.S. Reissue Pat. No. 28,353 and wind the web on a core which is rotated by a mandrel. Surface rewinders are described, for example, in U.S. Pat. Nos. 4,723,724 and 5,104,055 and wind the web on a core which is rotated by a three roll cradle.
The critical operation in both center rewinders and surface rewinders is the sequence of steps referred to as cutoff and transfer. The web must be severed to end the winding of one roll, the leading edge of the severed web must be transferred to a new core, and the new core must be rotated to begin winding a new roll. These steps must be accomplished repeatedly and reliably while the web is moving at high speed. It is also desirable that each roll have exact sheet count and that the web is wound uniformly and substantially without wrinkles.
In U.S. Pat. No. 4,723,724 a stationary plate or dead plate (217 in FIGS. 11-15; 317 in FIG. 18; 417 in FIG. 18A) upstream of the second winding roll is used to initiate core rotation and to transfer the web to a glue-equipped core. The core pinches the web against the stationary plate to tension and sever the web, and the web is wound on the core as the core rolls along the stationary plate. In FIGS. 11-15 a rotating pinch arm 221 presses the web against an upper belt 209 to isolate a line of perforations P on which the web is severed.
U.S. Pat. No. 5,137,225 also describes a surface rewinder which uses a stationary surface to effect a temporary braking of the web between the stationary surface and the core, thus causing a tearing of the web between the just-finished roll and the incoming core. This process, which uses the core to pinch and slow down the web, stretches the web from the pinch point of the core to the finished wound roll to snap a perforation between the two points. This long distance between the core and the finished roll must be elongated by at least the percentage of stretch in the material, commonly 6 to 25%. This elongation is created by the core being pinched to the stationary surface with the core insertion speed being less than the web speed. In effect, there is at least the same amount of slack web generated upstream of the inserted core as is required to elongate and break the web downstream of the core, plus the distance the core must still travel before it reaches the first winding roll and is accelerated to web speed.
The problems with this method are the significant amount of slack web generated upstream, and the difficulty in running short perforations which result in more than one perforation between the inserted core and the finished wound roll. The excess generated slack causes uncontrollable wrinkling and web tension problems which limit the speed of the machine. The long distance from the core to the finished wound roll also limits the length of perforation which can be run, and the maximum stretch which can be run. This method also requires a stiff core to pinch the web to the stationary surface to minimize slippage of the web as it is stretched, thus increasing the cost of the cores.
European Patent 0 694 020B1 and U.S. Pat. No. 5,979,818 use a pad/presser member to cooperate with surface portions of the first winding roll which have a low coefficient of friction. This low coefficient of friction on the first winding roll is highly undesirable as it permits winding products to become unstable during winding due to slippage between the product and the winding drums. This is explained in U.S. Pat. Nos. 5,370,335 and 5,505,405.